A method, device, medium and equipment for constructing a failure case knowledge management system

Abstract

The embodiments of the present application disclose a method, apparatus, medium and equipment for constructing a failure case knowledge management system, which relates to the field of data management technology, including: performing knowledge conversion on text data to obtain failure keywords; performing embedding construction on parameter signal data to obtain embedded data; generating rules based on the embedded data using a large language model to obtain failure rules; and constructing a failure knowledge network based on failure keywords and failure rules to realize the construction of a knowledge management system. Since the present application introduces embedding technology in the conversion, it can realize the non-aligned knowledge conversion of multimodal data of traditional text data and parameter signal data, and the embedding technology includes three embedding means to convert important time series information such as the relationship and time series of multi-dimensional monitoring parameters into knowledge, while retaining the time series characteristics, improving the learning ability of the knowledge process for the time relationship in the failure process, and improving the management effect of the knowledge management system of failure cases.

Description

Technical Field

[0001] The present application relates to the field of data management technology, and in particular to a method, apparatus, medium and equipment for constructing a failure case knowledge management system. Background Art

[0002] In the aviation field, various failure cases and fault data are highly correlated. The complex structure of aviation machinery and the information exchange between various modules lead to a wide variety of fault phenomena and types. Due to the close connections between equipment, there is a many-to-many mapping between fault phenomena and causes; that is, the same cause can lead to multiple fault manifestations, or the same fault phenomenon can be caused by different causes. However, existing failure case information management systems are simple in structure and cannot efficiently correlate and effectively manage these complex relationships. Summary of the Invention

[0003] The main purpose of this application is to provide a method, device, medium and equipment for constructing a failure case knowledge management system, aiming to solve the problem of poor management effect of failure case information management in the existing technology.

[0004] To achieve the above objectives, the technical solutions adopted in the embodiments of the present application are as follows:

[0005] In a first aspect, an embodiment of the present application provides a method for constructing a failure case knowledge management system, comprising the following steps:

[0006] Perform knowledge conversion on text data to obtain invalid keywords;

[0007] Embedding the parameter signal data to obtain embedded data; wherein the embedding construction includes absolute position embedding, temporal causal embedding, and multi-parameter alignment embedding;

[0008] Based on the embedded data, a large language model is used to generate rules and obtain failure rules;

[0009] According to the failure keywords and failure rules, a failure knowledge network is constructed to realize the construction of a knowledge management system.

[0010] In a possible implementation of the first aspect, embedding the parameter signal data to obtain the embedded data includes:

[0011] The parameter signal data is embedded and constructed, and the absolute position embedding, temporal causal embedding and multi-parameter alignment embedding of the parameter signal data are calculated respectively. The calculated multiple embeddings are added to the parameter signal data to obtain the embedded data.

[0012] In a possible implementation of the first aspect, before constructing the expiration knowledge network based on the expiration keywords and the expiration rules to implement the construction of the knowledge management system, the method further includes:

[0013] Obtain failure semantic units based on failure keywords and embedded data;

[0014] Based on failure keywords and failure rules, a failure knowledge network is constructed to realize the construction of a knowledge management system, including:

[0015] According to the invalidation keywords and invalidation rules, a graph network structure is used as the formal carrier, invalidation semantic units are used as nodes of the graph network, and invalidation relations are used as edges of the graph network. An invalidation knowledge network is constructed to realize the construction of a knowledge management system.

[0016] In a possible implementation of the first aspect, after constructing an expiration knowledge network based on expiration keywords and expiration rules to implement the construction of a knowledge management system, the method further includes:

[0017] Purify the failure rules to obtain the target failure rules;

[0018] The target failure rules are mapped back to the failure knowledge network to obtain the target failure knowledge network.

[0019] In a possible implementation of the first aspect, purifying the invalidation rules to obtain target invalidation rules includes:

[0020] Sampling and extracting rules from the failure knowledge network to form a rule set;

[0021] After dividing the rule set into nodes, rule calculation is performed based on the divided nodes to obtain the target failure rules.

[0022] In a possible implementation of the first aspect, after constructing an expiration knowledge network based on expiration keywords and expiration rules to implement the construction of a knowledge management system, the method further includes:

[0023] Merge rules based on target failure cases to update failure rules;

[0024] Based on the updated failure rules, the failure knowledge network is updated.

[0025] In a possible implementation of the first aspect, before performing knowledge conversion on the text data to obtain invalid keywords, the method further includes:

[0026] Based on historical failure cases, text data and parameter signal data are obtained.

[0027] In a second aspect, an embodiment of the present application provides a device for constructing a failure case knowledge management system, including:

[0028] The conversion module is used to convert text data into knowledge and obtain invalid keywords;

[0029] The embedding module is used to embed parameter signal data to obtain embedded data; wherein, the embedding construction includes absolute position embedding, temporal causal embedding, and multi-parameter alignment embedding;

[0030] The generation module is used to generate rules based on the embedded data using a large language model to obtain invalidation rules;

[0031] Construction module,The construction module is used to construct a failure knowledge network based on failure keywords and failure rules to realize the construction of a knowledge management system.

[0032] In a third aspect, an embodiment of the present application provides a computer-readable storage medium storing a computer program. When the computer program is loaded and executed by a processor, it implements the method for constructing a failure case knowledge management system as provided in any one of the first aspects above.

[0033] In a fourth aspect, an embodiment of the present application provides an electronic device, including a processor and a memory, wherein:

[0034] Memory is used to store computer programs;

[0035] The processor is used to load and execute a computer program so that the electronic device executes the method for constructing a failure case knowledge management system as provided in any one of the first aspects above.

[0036] Compared with the prior art, the present invention has the following advantages:

[0037] The present application proposes a method, device, medium and equipment for constructing a failure case knowledge management system. The method includes: performing knowledge conversion on text data to obtain failure keywords; performing embedding construction on parameter signal data to obtain embedded data; wherein the embedding construction includes absolute position embedding, temporal causal embedding and multi-parameter alignment embedding; based on the embedded data, using a large language model to generate rules to obtain failure rules; and constructing a failure knowledge network based on the failure keywords and failure rules to realize the construction of the knowledge management system. The present application converts text data and parameter signal data into knowledge to obtain failure keywords and embedded data to generate failure rules for constructing a failure knowledge network. Since embedding technology is introduced in the conversion, it can realize the non-aligned knowledge conversion of multimodal data of traditional text data and parameter signal data, and the embedding technology includes absolute position embedding, temporal causal embedding and multi-parameter alignment embedding, which converts the relationship and time sequence of multi-dimensional monitoring parameters into knowledge, and improves the learning ability of the knowledge process for the time relationship in the failure process while retaining the time sequence characteristics, thereby improving the management effect of the knowledge management system of failure cases. BRIEF DESCRIPTION OF THE DRAWINGS

[0038] Figure 1 A schematic diagram of the electronic device structure of the hardware operating environment involved in the embodiments of the present application;

[0039] Figure 2 A flowchart of a method for constructing a failure case knowledge management system provided in an embodiment of the present application;

[0040] Figure 3 A flowchart of a method for constructing a failure case knowledge management system according to an embodiment of the present application is provided;

[0041] Figure 4 A schematic diagram of modules of a device for constructing a failure case knowledge management system according to an embodiment of the present application;

[0042] Markings in the figure: 101 - processor, 102 - communication bus, 103 - network interface, 104 - user interface, 105 - memory. DETAILED DESCRIPTION

[0043] It should be understood that the specific embodiments described herein are only used to explain the present application and are not intended to limit the present application.

[0044] The main solutions of the embodiments of the present application are: performing knowledge conversion on text data to obtain failure keywords; performing embedding construction on parameter signal data to obtain embedded data; wherein the embedding construction includes absolute position embedding, temporal causal embedding and multi-parameter alignment embedding; based on the embedded data, using a large language model to generate rules to obtain failure rules; based on the failure keywords and failure rules, constructing a failure knowledge network to realize the construction of a knowledge management system.

[0045] In the aviation field, various failure cases and fault data are highly correlated. The complex structure of aviation machinery, coupled with the interactive information exchange between various modules, leads to a wide variety of fault phenomena and types. Due to the close interconnectedness of equipment, there is a many-to-many mapping between fault phenomena and causes. The same fault cause can lead to multiple fault manifestations, and the same fault phenomenon can be caused by different causes. However, existing failure case information management systems are simple in structure and cannot efficiently correlate and effectively manage these complex relationships.

[0046] Furthermore, after a device fails, its information is primarily recorded through two methods: sensor parameters and staff notes. However, due to a lack of alignment between sensor parameters and records, existing systems are only capable of organizing and managing single data sets, making it impossible to comprehensively analyze equipment failures during failure information research. In particular, existing failure case management systems require manual data entry, preventing failure recorders from subjectively recording data rich in timing parameter information about the failure process. This results in isolated islands of failure information between different modalities, significantly complicating failure case management.

[0047] All of the above problems indicate that the existing failure case management system is not sufficient to achieve effective management of failure knowledge, which seriously hinders the development of aviation intelligence.

[0048] Therefore, the present application provides a solution, which obtains failure keywords and embedded data through knowledge-based conversion of text data and parameter signal data to generate failure rules for constructing a failure knowledge network. Since embedding technology is introduced in the conversion, it can realize the non-aligned knowledge-based conversion of multimodal data of traditional text data and parameter signal data, and the embedding technology includes absolute position embedding, time series causal embedding and multi-parameter alignment embedding, which converts the relationship, time series and other important time series information of multi-dimensional monitoring parameters into knowledge, while retaining the time series characteristics, improving the learning ability of the knowledge-based process for the time relationship in the failure process, thereby improving the management effect of the knowledge management system of failure cases.

[0049] Refer to the attached Figure 1 , attached Figure 1This is a schematic diagram of the structure of an electronic device of the hardware operating environment involved in the embodiment of the present application. The electronic device may include: a processor 101, such as a central processing unit (CPU), a communication bus 102, a user interface 104, a network interface 103, and a memory 105. Among them, the communication bus 102 is used to realize the connection and communication between these components. The user interface 104 may include a display screen (Display), an input unit such as a keyboard (Keyboard), and optionally the user interface 104 may also include a standard wired interface and a wireless interface. The network interface 103 may optionally include a standard wired interface and a wireless interface (such as a wireless fidelity (WIreless-FIdelity, WI-FI) interface). The memory 105 may optionally be a storage device independent of the aforementioned processor 101. The memory 105 may be a high-speed random access memory (RAM) memory, or a stable non-volatile memory (NVM), such as at least one disk storage. The processor 101 may be a general-purpose processor, including a central processing unit, a network processor, etc., or may be a digital signal processor, an application-specific integrated circuit, a field programmable gate array or other programmable logic device, a discrete gate or transistor logic device, or a discrete hardware component.

[0050] Those skilled in the art will understand that Figure 1 The structure shown in the figure does not constitute a limitation to the electronic device, and may include more or fewer components than shown in the figure, or combine certain components, or arrange the components differently.

[0051] As attached Figure 1 As shown, the memory 105 as a storage medium may include an operating system, a network communication module, a user interface module, and a failure case knowledge management system construction device.

[0052] In the attached Figure 1 In the electronic device shown, the network interface 103 is mainly used for data communication with the network server; the user interface 104 is mainly used for data interaction with the user; the processor 101 and the memory 105 in this application can be set in the electronic device, and the electronic device calls the failure case knowledge management system construction device stored in the memory 105 through the processor 101, and executes the failure case knowledge management system construction method provided in the embodiment of this application.

[0053] Refer to the attached Figure 2 Based on the hardware device of the aforementioned embodiment, the embodiment of the present application provides a method for constructing a failure case knowledge management system, comprising the following steps:

[0054] S10: Perform knowledge conversion on the text data to obtain invalid keywords.

[0055] During the specific implementation process, existing text data is transformed into knowledge, and words or phrases related to the failure topic are automatically extracted from the text. These keywords can summarize the subject and content of the failure record. Failure semantic extraction methods mainly include failure entity extraction, failure relationship extraction, failure time extraction, and failure location extraction. Failure keywords are extracted through keyword extraction algorithms, including but not limited to the TF-IDF algorithm, TextRank algorithm, LSA algorithm, and LDA algorithm, to achieve a semantic mapping representation of the data. Knowledge transformation is to encode knowledge in a certain form and structure for easy storage, dissemination, and sharing. Encoded knowledge can be in the form of text, images, audio, video, etc., or it can be structured data and information. Encoded knowledge can be disseminated and shared through digitalization, networking, and other technical means, making knowledge acquisition and utilization more convenient and efficient.

[0056] S20: Embedding and constructing the parameter signal data to obtain embedded data; wherein the embedding and constructing includes absolute position embedding, temporal causal embedding, and multi-parameter alignment embedding.

[0057] In the specific implementation process, embedding and constructing parameter signal data is actually converting it into knowledge. Through rule-oriented time series embedding technology, combined with the technology of logical rules in time series data and knowledge graph, entities, relationships and their information changing over time are embedded and represented, which can capture and utilize the time dependency between entities, thereby providing deeper data analysis and more accurate predictions. Among them, absolute position embedding provides the positioning of time series in the overall data collection process, time series causal embedding provides the representation of the correlation between parameters, and multi-parameter alignment embedding is used to provide the expression of the correlation between parameters; the absolute position embedding, time series causal embedding, and multi-parameter alignment embedding of the data sample are calculated respectively, and the embedding is added to the data, and finally, a multi-angle information description of the failure degradation data is formed. That is: embedding and constructing the parameter signal data to obtain embedded data, including:

[0058] The parameter signal data is embedded and constructed, and the absolute position embedding, temporal causal embedding and multi-parameter alignment embedding of the parameter signal data are calculated respectively. The calculated multiple embeddings are added to the parameter signal data to obtain the embedded data.

[0059] Specifically: Absolute position embedding, used to describe fixed positions in parameters, conveys position information by adding a fixed vector representation to each position in the sequence. These vectors capture the absolute position of each position in the sequence and are relatively independent of its relationship with other positions.

[0060] Temporal causal embedding is used to capture the temporal and causal relationships in sequence data by encoding the temporal and causal relationships into vector representations and embedding them along with the sequence data.

[0061] Multi-parameter alignment embedding aligns multiple parameters or features from different data sources into a common embedding space, ensuring that similar features have similar representations within this space. This is because in real-world failure environments, the collected data comes from multiple parameters, which exhibit complex correlations due to the system's cross-linking relationships. These parameters collaboratively represent the system's current state.

[0062] In one embodiment, before performing knowledge conversion on text data to obtain invalid keywords, the method further includes:

[0063] Based on historical failure cases, text data and parameter signal data are obtained.

[0064] In the specific implementation process, the acquisition of text data and parameter signal data can rely on existing failure cases, that is, historical failure cases, that is, some failure cases generated in historical work in the aviation field, such as equipment failures.

[0065] S30: Generate rules using a large language model based on the embedded data to obtain invalidation rules.

[0066] During implementation, rule-based time-series embedding technology is used to transform important information, such as the relationships and time series of multi-dimensional monitoring parameters, into knowledge. Large language models are used to automatically generate rules. These models include but are not limited to GPT-3, BERT, and Megatron, which have a large number of parameters, can process large amounts of text data, and possess strong language understanding and generation capabilities.

[0067] S40: Based on the failure keywords and failure rules, a failure knowledge network is constructed to realize the construction of a knowledge management system.

[0068] In the specific implementation process, based on the semantic representation extracted in the above steps, i.e., the invalidation keywords, and the generated invalidation rules, which may include rule nodes and invalidation relationships, a knowledge network is constructed, using the graph network structure as a formal carrier, invalidation semantic units as nodes of the graph network, and invalidation relationships as edges of the graph network to construct an invalidation knowledge network. The invalidation knowledge network can query the internal nodes of the graph network by retrieving information such as entity names, relationship names, and entity attributes. The invalidation semantic units are formed through the knowledge process of data, i.e., before constructing the invalidation knowledge network based on invalidation keywords and invalidation rules to realize the construction of the knowledge management system, the method also includes:

[0069] Obtain failure semantic units based on failure keywords and embedded data;

[0070] Based on failure keywords and failure rules, a failure knowledge network is constructed to realize the construction of a knowledge management system, including:

[0071] According to the invalidation keywords and invalidation rules, a graph network structure is used as the formal carrier, invalidation semantic units are used as nodes of the graph network, and invalidation relations are used as edges of the graph network. An invalidation knowledge network is constructed to realize the construction of a knowledge management system.

[0072] In the specific implementation process, the failure semantic units extracted by the knowledge conversion process serve as the constituent entities of the failure knowledge network. At the same time, the failure semantic units can also form failure rules in the form of a tree structure. The knowledge network construction is based on the network organization of the semantic units formed by the knowledge conversion.

[0073] In this embodiment, through the knowledge conversion of text data and parameter signal data, failure keywords and embedded data are obtained to generate failure rules to construct a failure knowledge network. Since embedding technology is introduced in the conversion, it is possible to achieve non-aligned knowledge conversion of multimodal data of traditional text data and parameter signal data. The embedding technology includes absolute position embedding, time series causal embedding and multi-parameter alignment embedding, which converts important time series information such as the relationship and time series of multi-dimensional monitoring parameters into knowledge, while retaining the time series characteristics, the knowledge process is improved to learn the time relationship in the failure process, thereby improving the management effect of the knowledge management system of failure cases.

[0074] In one embodiment, after constructing an invalidation knowledge network based on invalidation keywords and invalidation rules to implement the construction of a knowledge management system, the method further includes:

[0075] Purify the failure rules to obtain the target failure rules;

[0076] The target failure rules are mapped back to the failure knowledge network to obtain the target failure knowledge network.

[0077] In the specific implementation process, as shown in the attached Figure 3 As shown in the figure, after obtaining the failure knowledge network, rule purification is performed to optimize the network. After sampling and extracting the rules in the knowledge network to form a rule set, rule optimization calculation is performed. Node division is first performed, and then rule optimization calculation is performed based on the nodes. Node evaluation is achieved by calculating node attributes to reduce the complexity of the knowledge network. In other words, the failure rules are purified to obtain the target failure rules, including:

[0078] Sampling and extracting rules from the failure knowledge network to form a rule set;

[0079] After dividing the rule set into nodes, rule calculation is performed based on the divided nodes to obtain the target failure rules.

[0080] During implementation, entities are used as rule heads in the knowledge network. Failure case information with rich semantic content is extracted based on inter-entity associations, thereby formulating the relationship logic between failure semantic units and implementing failure relationship descriptions. The process involves traversing the failure entity nodes in the network and composing failure rules based on each node, the nodes within the local scope of the node in the failure knowledge network, and the relationships between nodes. Node partitioning uses different failure semantic units as nodes and failure relationships as edges. Failure rules with an in-degree greater than 1 are defined as large- or medium-grained semantic relationships, while those with an in-degree equal to 1 are defined as small-grained relationships. By decomposing large-grained rules into small-grained ones, the optimization and parallelization of multi-grained rules is achieved.

[0081] Rule calculation is used for optimization, also known as rule calculation optimization, and can be categorized into two types: rule merging-based optimization and rule cost replacement-based optimization. Rule merging-based optimization combines numerous repetitive calculations, reducing the number of calculations, thereby reducing overall computation time and improving efficiency. Rule cost replacement-based optimization replaces high-cost rule calculation modules with low-cost ones, achieving optimization and improving computational efficiency.

[0082] In one embodiment, after constructing an invalidation knowledge network based on invalidation keywords and invalidation rules to implement the construction of a knowledge management system, the method further includes:

[0083] Merge rules based on target failure cases to update failure rules;

[0084] Based on the updated failure rules, the failure knowledge network is updated.

[0085] In the specific implementation process, as shown in the attached Figure 3 As shown in the figure, after the knowledge network is established, in order to realize the update of the knowledge network, the rules are merged for the target failure cases. The target failure cases are also the newly added failure cases. According to these cases, knowledge conversion is performed and the corresponding failure semantic units are generated and added. Then, the corresponding rules are merged to update the failure rules, thereby completing the update of the failure knowledge network.

[0086] Refer to the attached Figure 4 Based on the same inventive concept as in the aforementioned embodiment, the embodiment of the present application further provides a device for constructing a failure case knowledge management system, comprising:

[0087] The conversion module is used to convert text data into knowledge and obtain invalid keywords;

[0088] The embedding module is used to embed parameter signal data to obtain embedded data; wherein, the embedding construction includes absolute position embedding, temporal causal embedding, and multi-parameter alignment embedding;

[0089] The generation module is used to generate rules based on the embedded data using a large language model to obtain invalidation rules;

[0090] Construction module,The construction module is used to construct a failure knowledge network based on failure keywords and failure rules to realize the construction of a knowledge management system.

[0091] Those skilled in the art should understand that the division of the various modules in the embodiment is merely a division of logical functions, and in actual application, they can be fully or partially integrated into one or more actual carriers, and these modules can all be implemented in the form of software called through a processing unit, or all be implemented in the form of hardware, or in the form of a combination of software and hardware. It should be noted that the modules in the failure case knowledge management system construction device in this embodiment correspond one-to-one to the steps in the failure case knowledge management system construction method in the aforementioned embodiment. Therefore, the specific implementation of this embodiment can refer to the implementation of the aforementioned failure case knowledge management system construction method, and will not be repeated here.

[0092] Based on the same inventive concept as in the aforementioned embodiment, an embodiment of the present application further provides a computer-readable storage medium storing a computer program. When the computer program is loaded and executed by a processor, the method for constructing a failure case knowledge management system as provided in the embodiment of the present application is implemented.

[0093] Based on the same inventive concept as in the above embodiment, an embodiment of the present application further provides an electronic device, including a processor and a memory, wherein:

[0094] Memory is used to store computer programs;

[0095] The processor is used to load and execute a computer program so that the electronic device executes the failure case knowledge management system construction method provided in the embodiment of the present application.

[0096] In some embodiments, the computer-readable storage medium may be a memory device such as FRAM, ROM, PROM, EPROM, EEPROM, flash memory, magnetic surface mount memory, optical disk, or CD-ROM; or various devices including any one or any combination of the above memories. The computer may be various computing devices including smart terminals and servers.

[0097] In some embodiments, executable instructions may be in the form of a program, software, software module, script, or code, written in any form of programming language (including compiled or interpreted languages, or declarative or procedural languages), and may be deployed in any form, including as a stand-alone program or as a module, component, subroutine, or other unit suitable for use in a computing environment.

[0098] As an example, executable instructions may, but need not, correspond to a file in a file system, may be stored as part of a file that stores other programs or data, such as in one or more scripts in a HyperText Markup Language (HTML) document, in a single file dedicated to the program in question, or in multiple coordinating files (e.g., files storing one or more modules, subroutines, or code portions).

[0099] By way of example, executable instructions may be deployed to be executed on one computing device, or on multiple computing devices at one site, or on multiple computing devices distributed across multiple sites and interconnected by a communication network.

[0100] It should be noted that, in this document, the terms "comprises," "includes," or any other variations thereof are intended to encompass non-exclusive inclusion, such that a process, method, article, or system comprising a series of elements includes not only those elements but also other elements not explicitly listed, or elements inherent to such process, method, article, or system. In the absence of further limitations, an element defined by the phrase "comprising a ..." does not exclude the presence of other identical elements in the process, method, article, or system comprising the element.

[0101] The serial numbers of the above embodiments of the present application are for description only and do not represent the advantages or disadvantages of the embodiments.

[0102] Through the description of the above embodiments, those skilled in the art can clearly understand that the above-mentioned embodiment methods can be implemented by means of software plus the necessary general hardware platform. Of course, they can also be implemented by hardware, but in many cases the former is a better embodiment. Based on this understanding, the technical solution of the present application, or the part that contributes to the prior art, can be embodied in the form of a software product. The computer software product is stored in a storage medium (such as a read-only memory / random access memory, a magnetic disk, or an optical disk), and includes a number of instructions for enabling a multimedia terminal device (which can be a mobile phone, a computer, a television receiver, or a network device, etc.) to execute the methods described in each embodiment of the present application.

[0103] In summary, the present application provides a method, device, medium and equipment for constructing a failure case knowledge management system, which includes: performing knowledge conversion on text data to obtain failure keywords; performing embedding construction on parameter signal data to obtain embedded data; wherein the embedding construction includes absolute position embedding, temporal causal embedding and multi-parameter alignment embedding; based on the embedded data, using a large language model to generate rules to obtain failure rules; based on the failure keywords and failure rules, constructing a failure knowledge network to realize the construction of a knowledge management system. The present application obtains failure keywords and embedded data to generate failure rules to construct a failure knowledge network through knowledge conversion of text data and parameter signal data. Since embedding technology is introduced in the conversion, it can realize the non-aligned knowledge conversion of multimodal data of traditional text data and parameter signal data, and the embedding technology includes absolute position embedding, temporal causal embedding and multi-parameter alignment embedding, which converts the relationship and time series of multi-dimensional monitoring parameters into knowledge, while retaining the time series characteristics, improving the learning ability of the knowledge process for the time relationship in the failure process, thereby improving the management effect of the knowledge management system of failure cases.

[0104] The above description is only a preferred embodiment of the present application and is not intended to limit the present application. Any modifications, equivalent replacements, improvements, etc. made within the spirit and principles of the present application should be included in the scope of protection of the present application.

Claims

1. A method for constructing a failure case knowledge management system, characterized in that: The following steps are involved: Perform knowledge conversion on text data to obtain invalid keywords; Embedding and constructing the parameter signal data to obtain embedded data; wherein the embedding and constructing includes absolute position embedding, temporal causal embedding, and multi-parameter alignment embedding; embedding and constructing the parameter signal data to obtain embedded data includes: Embedding the parameter signal data, respectively calculating the absolute position embedding, the temporal causal embedding, and the multi-parameter alignment embedding of the parameter signal data, and adding the calculated multiple embeddings to the parameter signal data to obtain embedded data; wherein the absolute position embedding is used to describe a fixed position in the parameter, the temporal causal embedding is used to capture the time and causal relationship in the sequence data, and the multi-parameter alignment embedding is used to align multiple parameters or features from different data sources into a common embedding space; Generate rules using a large language model based on the embedded data to obtain failure rules; According to the failure keywords and the failure rules, a failure knowledge network is constructed to realize the construction of a knowledge management system.

2. The method for constructing a failure case knowledge management system according to claim 1, characterized in that: Before constructing the failure knowledge network based on the failure keywords and the failure rules to realize the construction of the knowledge management system, the method further includes: Obtaining invalid semantic units respectively according to the invalid keywords and the embedded data; The construction of the failure knowledge network based on the failure keywords and the failure rules to realize the construction of the knowledge management system includes: According to the invalidation keywords and the invalidation rules, a graph network structure is used as a formal carrier, the invalidation semantic units are used as nodes of the graph network, and the invalidation relations are used as edges of the graph network, so as to construct an invalidation knowledge network to realize the construction of a knowledge management system.

3. The method for constructing a failure case knowledge management system according to claim 1, characterized in that: After constructing the failure knowledge network based on the failure keywords and the failure rules to realize the construction of the knowledge management system, the method further includes: Purifying the failure rules to obtain target failure rules; The target failure rules are mapped back to the failure knowledge network to obtain a target failure knowledge network.

4. The method for constructing a failure case knowledge management system according to claim 3, characterized in that: The step of purifying the invalidation rules to obtain target invalidation rules includes: Sampling and extracting rules in the failure knowledge network to form a rule set; After dividing the rule set into nodes, rule calculation is performed based on the divided nodes to obtain target failure rules.

5. The method for constructing a failure case knowledge management system according to claim 1, characterized in that: After constructing the failure knowledge network based on the failure keywords and the failure rules to realize the construction of the knowledge management system, the method further includes: performing rule merging according to target failure cases to update the failure rules; The failure knowledge network is updated based on the updated failure rules.

6. The method for constructing a failure case knowledge management system according to claim 1, characterized in that: Before performing knowledge conversion on the text data to obtain invalid keywords, the method further includes: The text data and the parameter signal data are obtained based on historical failure cases.

7. A device for constructing a failure case knowledge management system, characterized in that: include: A conversion module, the conversion module is used to perform knowledge conversion on text data to obtain invalid keywords; An embedding module is used to embed parameter signal data to obtain embedded data; wherein the embedding construction includes absolute position embedding, temporal causal embedding, and multi-parameter alignment embedding; the embedding construction of parameter signal data to obtain embedded data includes: Embedding the parameter signal data, respectively calculating the absolute position embedding, the temporal causal embedding, and the multi-parameter alignment embedding of the parameter signal data, and adding the calculated multiple embeddings to the parameter signal data to obtain embedded data; wherein the absolute position embedding is used to describe a fixed position in the parameter, the temporal causal embedding is used to capture the time and causal relationship in the sequence data, and the multi-parameter alignment embedding is used to align multiple parameters or features from different data sources into a common embedding space; A generation module, configured to generate rules based on the embedded data using a large language model to obtain failure rules; A construction module is used to construct an expiration knowledge network according to the expiration keywords and the expiration rules to realize the construction of a knowledge management system.

8. A computer-readable storage medium storing a computer program, characterized in that: When the computer program is loaded and executed by a processor, the method for constructing a failure case knowledge management system according to any one of claims 1 to 6 is implemented.

9. An electronic device, characterized in that: comprising a processor and a memory, wherein: The memory is used to store computer programs; The processor is configured to load and execute the computer program so as to enable the electronic device to execute the method for constructing a failure case knowledge management system according to any one of claims 1 to 6.

Images